Toward a more robust facial expression recognition in occluded images using randomly sampled Gabor based templates

Occlusion is a big challenge for facial expression recognition (FER) in real-world situations. Previous FER efforts to address occlusion suffer from loss of appearance features and are largely limited to a few occlusion types and single testing strategy. This paper presents a robust approach for FER in occluded images and addresses these issues. A set of Gabor based templates is extracted from images in the gallery using a Monte Carlo algorithm. These templates are converted into distance features using template matching. The resulting feature vectors are robust to occlusion. Occluded eyes and mouth regions and randomly places occlusion patches are used for testing. Two testing strategies analyze the effects of these occlusions on the overall recognition performance as well as each facial expression. Experimental results on the Cohn-Kanade database confirm the high robustness of our approach and provide useful insights about the effects of occlusion on FER. Performance is also compared with previous approaches.

Robust face recognition with partial occlusion, illumination variation and limited training data by optimal feature selection

This study investigates face recognition with partial occlusion, illumination variation and their combination, assuming no prior information about the mismatch, and limited training data for each person. The authors extend their previous posterior union model (PUM) to give a new method capable of dealing with all these problems. PUM is an approach for selecting the optimal local image features for recognition to improve robustness to partial occlusion. The extension is in two stages. First, authors extend PUM from a probability-based formulation to a similarity-based formulation, so that it operates with as little as one single training sample to offer robustness to partial occlusion. Second, they extend this new formulation to make it robust to illumination variation, and to combined illumination variation and partial occlusion, by a novel combination of multicondition relighting and optimal feature selection. To evaluate the new methods, a number of databases with various simulated and realistic occlusion/illumination mismatches have been used. The results have demonstrated the improved robustness of the new methods.

Largest Matching Areas for Illumination and Occlusion Robust Face Recognition

In this paper, we introduce a novel approach to face recognition which simultaneously tackles three combined challenges: 1) uneven illumination; 2) partial occlusion; and 3) limited training data. The new approach performs lighting normalization, occlusion de-emphasis and finally face recognition, based on finding the largest matching area (LMA) at each point on the face, as opposed to traditional fixed-size local area-based approaches. Robustness is achieved with novel approaches for feature extraction, LMA-based face image comparison and unseen data modeling. On the extended YaleB and AR face databases for face identification, our method using only a single training image per person, outperforms other methods using a single training image, and matches or exceeds methods which require multiple training images. On the labeled faces in the wild face verification database, our method outperforms comparable unsupervised methods. We also show that the new method performs competitively even when the training images are corrupted.

Random Gabor based templates for facial expression recognition in images with facial occlusion

Robust facial expression recognition (FER) under occluded face conditions is challenging. It requires robust algorithms of feature extraction and investigations into the effects of different types of occlusion on the recognition performance to gain insight. Previous FER studies in this area have been limited. They have spanned recovery strategies for loss of local texture information and testing limited to only a few types of occlusion and predominantly a matched train-test strategy. This paper proposes a robust approach that employs a Monte Carlo algorithm to extract a set of Gabor based part-face templates from gallery images and converts these templates into template match distance features. The resulting feature vectors are robust to occlusion because occluded parts are covered by some but not all of the random templates. The method is evaluated using facial images with occluded regions around the eyes and the mouth, randomly placed occlusion patches of different sizes, and near-realistic occlusion of eyes with clear and solid glasses. Both matched and mis-matched train and test strategies are adopted to analyze the effects of such occlusion. Overall recognition performance and the performance for each facial expression are investigated. Experimental results on the Cohn-Kanade and JAFFE databases demonstrate the high robustness and fast processing speed of our approach, and provide useful insight into the effects of occlusion on FER. The results on the parameter sensitivity demonstrate a certain level of robustness of the approach to changes in the orientation and scale of Gabor filters, the size of templates, and occlusions ratios. Performance comparisons with previous approaches show that the proposed method is more robust to occlusion with lower reductions in accuracy from occlusion of eyes or mouth.

Facial Feature Tracking and Occlusion Recovery in American Sign Language

Facial features play an important role in expressing grammatical information in signed languages, including American Sign Language(ASL). Gestures such as raising or furrowing the eyebrows are key indicators of constructions such as yes-no questions. Periodic head movements (nods and shakes) are also an essential part of the expression of syntactic information, such as negation (associated with a side-to-side headshake). Therefore, identification of these facial gestures is essential to sign language recognition. One problem with detection of such grammatical indicators is occlusion recovery. If the signer's hand blocks his/her eyebrows during production of a sign, it becomes difficult to track the eyebrows. We have developed a system to detect such grammatical markers in ASL that recovers promptly from occlusion. Our system detects and tracks evolving templates of facial features, which are based on an anthropometric face model, and interprets the geometric relationships of these templates to identify grammatical markers. It was tested on a variety of ASL sentences signed by various Deaf native signers and detected facial gestures used to express grammatical information, such as raised and furrowed eyebrows as well as headshakes.

Robust face recognition using posterior union model based neural networks

Face recognition with unknown, partial distortion and occlusion is a practical problem, and has a wide range of applications, including security and multimedia information retrieval. The authors present a new approach to face recognition subject to unknown, partial distortion and occlusion. The new approach is based on a probabilistic decision-based neural network, enhanced by a statistical method called the posterior union model (PUM). PUM is an approach for ignoring severely mismatched local features and focusing the recognition mainly on the reliable local features. It thereby improves the robustness while assuming no prior information about the corruption. We call the new approach the posterior union decision-based neural network (PUDBNN). The new PUDBNN model has been evaluated on three face image databases (XM2VTS, AT&T and AR) using testing images subjected to various types of simulated and realistic partial distortion and occlusion. The new system has been compared to other approaches and has demonstrated improved performance.

3D Morphable Model Construction for Robust Ear and Face Recognition

Recent work suggests that the human ear varies significantly between different subjects and can be used for identification. In principle, therefore, using ears in addition to the face within a recognition system could improve accuracy and robustness, particularly for non-frontal views. The paper describes work that investigates this hypothesis using an approach based on the construction of a 3D morphable model of the head and ear. One issue with creating a model that includes the ear is that existing training datasets contain noise and partial occlusion. Rather than exclude these regions manually, a classifier has been developed which automates this process. When combined with a robust registration algorithm the resulting system enables full head morphable models to be constructed efficiently using less constrained datasets. The algorithm has been evaluated using registration consistency, model coverage and minimalism metrics, which together demonstrate the accuracy of the approach. To make it easier to build on this work, the source code has been made available online.

Robust bimodal person identification using face and speech with limited training data and corruption of both modalities

This paper presents a novel method of audio-visual fusion for person identification where both the speech and facial modalities may be corrupted, and there is a lack of prior knowledge about the corruption. Furthermore, we assume there is a limited amount of training data for each modality (e.g., a short training speech segment and a single training facial image for each person). A new representation and a modified cosine similarity are introduced for combining and comparing bimodal features with limited training data as well as vastly differing data rates and feature sizes. Optimal feature selection and multicondition training are used to reduce the mismatch between training and testing, thereby making the system robust to unknown bimodal corruption. Experiments have been carried out on a bimodal data set created from the SPIDRE and AR databases with variable noise corruption of speech and occlusion in the face images. The new method has demonstrated improved recognition accuracy.

New approaches to dental occlusion: A literature update

Because the study of occlusion is a basic area in dentistry, its components, physiology and integration with the stomatognathic system (SS) have been the subject of interest in the scientific literature. However, the focus given to this issue has changed substantially. Currently, new approaches have been proposed in order to update concepts and to demonstrate the full integration and functionality of this system within the human body. With this approach, the authors proposed the following literature review aimed at gathering recent papers (published from 2000 to 2010) with innovative study design, methodology and/or results. The authors' intention is to show the main trends in the study of occlusion and the SS. The literature review was conducted in the PubMed database, using initially the term dental occlusion as a key-word. As items of interest were found, papers were grouped by categories according to their main subject matter. Forty-seven articles were selected and the main categories obtained were: 1. functional magnetic resonance imaging (fMRI); 2. brain activation; 3. masticatory/occlusal function; 4. body function and physical performance; 5. osseoperception; 6. finite element models; and 7. occlusion and pain. Observing the current literature, the authors found that recent studies present different methodologies for the study of occlusion. These studies have allowed scientists to obtain detailed information about the physiology of occlusion and the SS, as well as about its integration in the body. Research in this area should be continued in order to clarify, in detail, the role of each component of the SS and its interaction with human physiology. © 2012 by CHROMA, Inc.

Facial type measurements influence on transverse dimensions of normal occlusion arches

Objective – To correlate facial type measurements of Caucasian individuals with transverse dimensions of normal occlusion arches. Methods – Twenty-one pairs of dental models were selected according to the following inclusion criteria: presence of all permanent teeth from 1 st molar to 1 st molar; normal occlusion; no prosthetic crowns; no previous orthodontic treatment and 2 mm or less of crow- dings or spacings. The cephalometric measurements of lateral cephalometric X-ray of the same individuals were taken and tabulat ed. To evaluate the repetition of arch measurements, paired Student’s t-test and Pearson's correlation coefficient were used. The r elationship between the measurements was analysed by using the Pearson’s correlation. Results – The repetition of the measurements showed high correlation and no systematic error. In the comparison between the measurements, a moderate negative correlation was observed b et- ween facial axis angle and the measurements Upper and Lower 6-6, whereas a positive correlation was observed between dentition height and the latter. Conclusion – It was observed a negative correlation between facial axis angle and upper and lower inter-molar distance as well as a positive correlation between dentition height and upper and lower inter-molar distance.

Relationship between the morphology of the maxillary central incisor and horizontal and vertical measurements of the face

Purpose: In order to assist in the selection of artificial teeth for complete dentures, this study aimed to assess the relationship between horizontal and vertical measurements of the face and the morphology of the maxillary central incisor. Materials and Methods: This was a study of 50 plaster casts and 100 teleradiographs - 50 in lateral norm and 50 in frontal norm, belonging to 50 individuals, Caucasian, with a naturally optimal occlusion, matching at least four of the six keys of Andrews. Images of the upper central incisors were obtained by scanning the plaster casts (three-dimensional) and subjectively classified by three examiners as oval, triangular or quadrangular. Facial measures (vertical and horizontal) were defined by means of teleradiographs. In order to check inter-examiner agreement on the classification of central incisor, the Kappa test was used. To verify whether data had normal distribution, the Kolmogorov-Smirnov test was used ( P > 0.2) was used. One-way analysis of variance was employed to assess the association between variables (P > 0.05). Results: When vertical measurements were compared with the three incisor shapes, there was no statistically significant difference (P > 0.05): Triangular (0.54), oval (0.63) and quadrangular (0.51). Similarly, no difference (P > 0.05) was found for facial width (139.08, 143.37, 141.65), maxillary width (76.68, 78.99, 76.91) and mandibular width (103.47, 105.50, 103.11). Conclusions: The majority of cases showed that horizontal and vertical measurements of the face cannot be used as a reference for determining the morphology of the maxillary central incisor crown. It is relevant to analyze and compare other morphological structures to improve the oral health-related quality of life for the conventional denture wearer.

Time window for recanalization in basilar artery occlusion: Speculative synthesis.

Basilar artery occlusion (BAO) is one of the most devastating forms of stroke and few patients have good outcomes without recanalization. Most centers apply recanalization therapies for BAO up to 12-24 hours after symptom onset, which is a substantially longer time window than the 4.5 hours used in anterior circulation stroke. In this speculative synthesis, we discuss recent advances in BAO treatment in order to understand why and under which circumstances longer symptom duration might not necrotize the brainstem and turn therapeutic attempts futile. We raise the possibility that distinct features of the posterior circulation, e.g., highly developed, persistent collateral arterial network, reverse filling of the distal basilar artery, and delicate plasma flow siding the clot, might sustain brittle patency of brainstem perforators in the face of stepwise growth of the thrombus. Meanwhile, the tissue clock characterizing the rapid necrosis of a typical anterior circulation penumbra will not start. During this perilous time period, recanalization at any point would salvage the brainstem from eventual necrosis caused by imminent reinforcement and further building up of the clot.

AVALIAÇÃO DO PADRÃO ESQUELÉTICO VERTICAL DA FACE NA OCLUSÃO NORMAL NATURAL E NAS MALOCLUSÕES DE ANGLE, E SUA CORRELAÇÃO COM A SÍNFISE MANDIBULAR

O propósito deste estudo foi avaliar cefalometricamente, o padrão esquelético vertical da face em indivíduos com oclusão normal natural e nas diferentes maloclusões e sua correlação com a sínfise mandibular, além de avaliar a presença de dimorfismo sexual. A amostra foi composta de 200 telerradiografias cefalométricas, divididas quanto ao tipo de oclusão, em cinco grupos: grupo A, com pacientes portadores de oclusão normal natural e grupos B, C, D e E, com pacientes portadores de maloclusões, sendo cada grupo, dividido igualmente quanto ao sexo e apresentando idade média entre 13 e 16 anos. A amostra foi classificada em 3 padrões morfológicos verticais da face, de acordo com o índice da altura facial (FHR), proposto por SIRIWAT & JARABAK ou Quociente de Jarabak, em: Hiperdivergente, Neutro e Hipodivergente. Foi utilizada a variável GoMe.VT, da análise de VIGORITO, para avaliar a inclinação da sínfise e sua correlação com os padrões verticais faciais. Após a coleta de dados e da avaliação dos testes estatísticos; qui-quadrado, teste t de Student e da correlação de Pearson, concluiu-se que, o padrão Hipodivergente em todos os pacientes estudados foi o mais frequente, com 70%, sendo que a maior frequência deste padrão foi encontrado na maloclusão Classe II, divisão 2, com 87.5%, existindo outras prevalências de alguns padrões em diferentes classes de oclusões. Foi encontrada uma correlação positiva entre a inclinação da sínfise mandibular e o quociente de Jarabak apenas para a maloclusão Classe I e maloclusão Classe III. Não houve diferença estatisticamente significante entre os sexos e a classificação da morfologia quando comparados os cinco grupos, porém, quando os grupos foram analisados separadamente, foram encontradas diferenças significantes entre os sexos.

Avaliação de medidas craniofaciais angulares e lineares em relação à quantidade de crescimento da face

A identificação dos padrões morfológicos da face é de extrema importância para a elaboração do diagnóstico ortodôntico e conseqüente plano de tratamento. Sendo assim, esta pesquisa teve como objetivo avaliar as relações entre as medidas lineares N-ENA, ENA-Pog e N-Pog, as medidas angulares N.OPI.ENA, ENA.OPI.Pog e N.OPI.Pog, e o índice VERT de acordo com a classificação proposta por RICKETTS. O material de estudo foi constituído de 700 telerradiografias, em norma lateral, de indivíduos brasileiros, leucodermas, de ambos os gêneros (318 do gênero masculino e 382 do gênero feminino), apresentando idade média de 16,66 anos. Das 700 telerradiografias foram constituídos quatro grupos de acordo com a oclusão. O primeiro grupo foi constituído de indivíduos com oclusão normal natural segundo ANDREWS, sendo observadas, clinicamente, quatro das seis chaves de oclusão. Os demais grupos se constituíram de indivíduos com más-oclusões de Classe I, II e III segundo ANGLE, sem relato de tratamento ortodôntico anterior. Após a análise estatística para a avaliação das medidas lineares N-ENA, ENA-Pog e N-Pog, e medidas angulares N.OPI.ENA, ENA.OPI.Pog e N.OPI.Pog, em relação ao índice VERT, evidenciou um aumento progressivo das medidas pesquisadas em relação aos seis padrões faciais determinados pelo índice VERT, mostrando um aumento destas medidas conforme se vai do padrão braquifacial severo para o padrão dolicofacial severo.